On Thu, Feb 7, 2013 at 11:01 AM, Jed Rothwell <[email protected]> wrote: > Jack Cole <[email protected]> wrote: > >> Seems to me like they could do something like that with a calibration run. >> Heat with the inactive wire, then put 10watts through the active wire. It >> should then show up as 10W excess if they leave that power input out of the >> calculation. > > > That's what calibrations are for! > > That's what they are. > > - Jed
Calibrations involve a method analysis. Daniel's point is that a method of analysis can be flawed if it generates a false positve signal OR if it masks a positive signal. The method analysis should be capable of detecting both a positive (desired) signal and a negative (null, undesired) signal. To test the method analsysis the system should be fed a "dummy" positive signal and "dummy" negative signal. Harry Harry
